> For the complete documentation index, see [llms.txt](https://docs.11x.ai/llms.txt).
> Use it to discover all available pages before exploring further.

# Audit and oversight

How to evidence what a digital worker did: call recordings and transcripts, call quality assessments, knowledge base version history, approval trails, and CRM write-back.


"What did the AI actually say?" is a question you will eventually be asked - by a rep, a customer, a security review, or your own legal team. 11x produces an audit trail across five surfaces: **call recordings and transcripts**, **call quality assessments**, **knowledge base version history**, **approval trails**, and **CRM write-back**.

<Callout kind="tip">
Set this up before you need it. The uncomfortable version of this question arrives after an incident, and by then the useful thing is a record that already exists rather than one you start building.
</Callout>

## The five audit surfaces

| Surface | What it evidences |
| --- | --- |
| **Call recording, transcription, summarization** | Every call, verbatim - what was said, promised, and disclosed |
| **Assessments and Smart Outcomes** | Call quality evaluation and structured outcomes at scale, so quality is auditable rather than assumed |
| **Knowledge base version history** | What changed, when, and by whom, with rollback |
| **Approval trail** | Who signed off on messaging, where an approval workflow is in use |
| **CRM write-back** | The durable record - interactions, transcripts, scores, and outcomes on the account |

## Why CRM write-back is the important one

The other surfaces live in 11x. CRM write-back puts the record where your reporting, your reps, and your retention policies already are - and where it survives any change to your 11x configuration.

Confirm transcripts, qualification results and outcomes land on the objects and fields your reporting expects, and fix mapping early. Backfilling misplaced CRM data later is painful and usually incomplete. See [security and data handling](/security/overview).

<Callout kind="alert">
Transcripts contain what a prospect actually said, verbatim - potentially commercial detail, personal circumstances, or information about third parties. Decide deliberately who can read them, and avoid piping full transcripts into systems with broad internal access.
</Callout>

## A practical oversight cadence

<Steps>
  <Step title="Weekly: sample auto-sent replies" icon="inbox">
    Especially objection-category replies, where a confident wrong answer does the most damage. See [smart replies](/alice/smart-replies).
  </Step>
  <Step title="Monthly: read disqualified transcripts" icon="file-search">
    This is where the expensive mistakes hide. A wrongly disqualified good lead disappears without anyone reporting it.
  </Step>
  <Step title="Monthly: collect flagged meetings from reps" icon="thumbs-down">
    Formally. Unflagged bad meetings mean thresholds stay wrong and trust erodes quietly.
  </Step>
  <Step title="On every change: use version history" icon="history">
    Change one thing at a time so a shift in results can be attributed.
  </Step>
  <Step title="Quarterly: audit access" icon="shield">
    Who holds admin rights, and whether any integration is still authorized with personal credentials.
  </Step>
</Steps>

## What to capture for a security review

Reviews tend to ask the same things. Have these ready:

- Which data each worker processes - see [security and data handling](/security/overview)
- Encryption in transit and at rest, and hosting
- Certifications: SOC 2 Type II, CASA Tier 3, GDPR, CCPA - see the [trust center](https://trust.11x.ai/)
- Whether models are trained on your data (they are not - data is used only in aggregate and anonymised form)
- Retention and subprocessors
- Access controls and CRM permission scoping
- Recording disclosure configuration

## Frequently Asked Questions

<Expandable title="Can we prove what a digital worker said on a specific call?">
Yes. Every call is recorded, transcribed, and summarized, and the record syncs to the CRM account.
</Expandable>

<Expandable title="How do we audit quality across thousands of calls?">
Assessments and Smart Outcomes evaluate call quality and record structured outcomes, so quality is measurable rather than sampled by hand. See [ICP scoring](/julian/icp-scoring).
</Expandable>

<Expandable title="How do we track what changed in our messaging?">
Knowledge base version history records what changed, when, and by whom, and supports rollback. See [knowledge base](/alice/knowledge-base).
</Expandable>

<Expandable title="Can we require human sign-off before messages send?">
Yes. Set an approval workflow so a named person approves before prospects see anything. See [brand protection](/security/brand-protection).
</Expandable>

<Expandable title="Does 11x train models on our data?">
No. Per the terms, data is used only in aggregate and anonymised form and AI models are not trained on confidential information. You retain ownership of what you submit.
</Expandable>

<Expandable title="Where should the durable record live?">
Your CRM. It survives configuration changes, sits inside your existing retention policy, and is visible to reps. Verify field mapping early.
</Expandable>

<Expandable title="Who at 11x can access our data?">
Access controls fall under the SOC 2 Type II and CASA Tier 3 scopes. Request the control documentation via the [trust center](https://trust.11x.ai/).
</Expandable>

## Next steps

<Columns cols={2}>
  <Card title="Brand protection" icon="shield-check" href="/security/brand-protection" horizontal>
    Controlling what gets said.
  </Card>
  <Card title="Security and data handling" icon="shield" href="/security/overview" horizontal>
    What is processed, and scoping.
  </Card>
  <Card title="Regional compliance" icon="globe" href="/security/regional-compliance" horizontal>
    Obligations by market.
  </Card>
  <Card title="Data and privacy FAQ" icon="circle-help" href="/help-center/faq/data-and-privacy" horizontal>
    Buyer-facing answers.
  </Card>
</Columns>

## Need help?

Email [support@11x.ai](mailto:support@11x.ai), or [book time with the team](https://www.11x.ai/demo).
